Martijn Wieling’s research while affiliated with University of Groningen and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (26)


Figure 1: A conceptual visualization of LVQ classifiers. Left: training examples (circles) and learned prototypes (stars) with colours as labels. Right: illustration of the Nearest Prototype Classifier (NPC) strategy with an uncoloured circle as an unseen example. Note that the dashed lines represent borders between classes and dotted lines denote the difference between an unseen example and prototypes.
An explainable approach to detect case law on housing and eviction issues within the HUDOC database
  • Preprint
  • File available

October 2024

·

16 Reads

·

Martijn Wieling

·

Case law is instrumental in shaping our understanding of human rights, including the right to adequate housing. The HUDOC database provides access to the textual content of case law from the European Court of Human Rights (ECtHR), along with some metadata. While this metadata includes valuable information, such as the application number and the articles addressed in a case, it often lacks detailed substantive insights, such as the specific issues a case covers. This underscores the need for detailed analysis to extract such information. However, given the size of the database - containing over 40,000 cases - an automated solution is essential. In this study, we focus on the right to adequate housing and aim to build models to detect cases related to housing and eviction issues. Our experiments show that the resulting models not only provide performance comparable to more sophisticated approaches but are also interpretable, offering explanations for their decisions by highlighting the most influential words. The application of these models led to the identification of new cases that were initially overlooked during data collection. This suggests that NLP approaches can be effectively applied to categorise case law based on the specific issues they address.

Download

Intergenerational Language Transmission of Frisian and Low Saxon in the Netherlands

October 2024

·

6 Reads

·

Martijn Bartelds

·

·

[...]

·

Martijn Wieling

An important mechanism for language maintenance is transmission from parents to their children. This mechanism is stronger for the regional language Frisian than it is for Low Saxon in the northern Netherlands. In this study, we assessed many variables potentially associated with parental language transmission for these regional languages. We analyzed questionnaire responses from around 25,000 Frisian and Low Saxon speakers participating in the Lifelines cohort study. Transmission was strongly associated with whether their children's other parent speaks the same language, and with the frequency of language use in different social contexts. Other important factors included language attitudes and the degree of urbanization of the respondent's neighborhood. Taken together, these findings suggest that language maintenance for Frisian and Low Saxon could potentially be bolstered by adequately stimulating positive language attitudes and the use of the language in different social contexts by both the government and smaller societal organizations.


Predicting citations in Dutch case law with natural language processing

June 2023

·

91 Reads

·

4 Citations

Artificial Intelligence and Law

With the ever-growing accessibility of case law online, it has become challenging to manually identify case law relevant to one’s legal issue. In the Netherlands, the planned increase in the online publication of case law is expected to exacerbate this challenge. In this paper, we tried to predict whether court decisions are cited by other courts or not after being published, thus in a way distinguishing between more and less authoritative cases. This type of system may be used to process the large amounts of available data by filtering out large quantities of non-authoritative decisions, thus helping legal practitioners and scholars to find relevant decisions more easily, and drastically reducing the time spent on preparation and analysis. For the Dutch Supreme Court, the match between our prediction and the actual data was relatively strong (with a Matthews Correlation Coefficient of 0.60). Our results were less successful for the Council of State and the district courts (MCC scores of 0.26 and 0.17, relatively). We also attempted to identify the most informative characteristics of a decision. We found that a completely explainable model, consisting only of handcrafted metadata features, performs almost as well as a less well-explainable system based on all text of the decision.


shows results for each model and task, grouped by pre-training language and model size. Highest overall performance is achieved by DeBERTaV3 large , an English model. In Section 5,
DUMB: A Benchmark for Smart Evaluation of Dutch Models

May 2023

·

54 Reads

We introduce the Dutch Model Benchmark: DUMB. The benchmark includes a diverse set of datasets for low-, medium- and high-resource tasks. The total set of eight tasks include three tasks that were previously not available in Dutch. Instead of relying on a mean score across tasks, we propose Relative Error Reduction (RER), which compares the DUMB performance of models to a strong baseline which can be referred to in the future even when assessing different sets of models. Through a comparison of 14 pre-trained models (mono- and multi-lingual, of varying sizes), we assess the internal consistency of the benchmark tasks, as well as the factors that likely enable high performance. Our results indicate that current Dutch monolingual models under-perform and suggest training larger Dutch models with other architectures and pre-training objectives. At present, the highest performance is achieved by DeBERTaV3 (large), XLM-R (large) and mDeBERTaV3 (base). In addition to highlighting best strategies for training larger Dutch models, DUMB will foster further research on Dutch. A public leaderboard is available at https://dumbench.nl.




Fig. 3. The trajectory of suicidal ideation by time and age during the COVID-19 pandemic. Shown are the results of interaction between time and age on reported suicidal ideation in our main sample of 36,106 participants (age 18-45). The left panels show the trajectory of suicidal ideation for specific ages over time. The right panels show the trajectory of specific time points across age. The legends at the top of the graph denoted colour coding of groups. The gray rectangles highlight the three different nationwide lockdowns in the Netherlands.
Longitudinal analyses of depression, anxiety, and suicidal ideation highlight greater prevalence in the northern Dutch population during the COVID-19 lockdowns

November 2022

·

41 Reads

·

8 Citations

Journal of Affective Disorders

Background The pandemic of the coronavirus disease 2019 (COVID-19) has led to an increased burden on mental health. Aims To investigate the development of major depressive disorder (MDD), generalized anxiety disorder (GAD), and suicidal ideation in the Netherlands during the first fifteen months of the pandemic and three nation-wide lockdowns. Method Participants of the Lifelines Cohort Study –a Dutch population-based sample-reported current symptoms of MDD and GAD, including suicidal ideation, according to DSM-IV criteria. Between March 2020 and June 2021, 36,106 participants (aged 18–96) filled out a total of 629,811 questionnaires across 23 time points. Trajectories over time were estimated using generalized additive models and analyzed in relation to age, sex, and lifetime history of MDD/GAD. Results We found non-linear trajectories for MDD and GAD with a higher number of symptoms and prevalence rates during periods of lockdown. The point prevalence of MDD and GAD peaked during the third hard lockdown at 2.88 % (95 % CI: 2.71 %–3.06 %) and 2.92 % (95 % CI: 2.76 %–3.08 %), respectively, in March 2021. Women, younger adults, and participants with a history of MDD/GAD reported significantly more symptoms. For suicidal ideation, we found a significant linear increase over time in younger participants. For example, 20-year-old participants reported 4.14× more suicidal ideation at the end of June 2021 compared to the start of the pandemic (4.64 % (CI: 3.09 %–6.96 %) versus 1.12 % (CI: 0.76 %–1.66 %)). Limitations Our findings should be interpreted in relation to the societal context of the Netherlands and the public health response of the Dutch government during the pandemic, which may be different in other regions in the world. Conclusions Our study showed greater prevalence of MDD and GAD during COVID-19 lockdowns and a continuing increase in suicidal thoughts among young adults suggesting that the pandemic and government enacted restrictions impacted mental health in the population. Our findings provide actionable insights on mental health in the population during the pandemic, which can guide policy makers and clinical care during future lockdowns and epi/pandemics.


Cognitive Benefits of Learning Additional Languages in Old Adulthood? Insights from an Intensive Longitudinal Intervention Study

August 2022

·

236 Reads

·

16 Citations

Applied Linguistics

Second language (L2) learning has been promoted as a promising intervention to stave off age-related cognitive decline. While previous studies based on mean trends showed inconclusive results, this study is the first to investigate nonlinear cognitive trajectories across a 30-week training period. German-speaking older participants (aged 64–75 years) enrolled for a Spanish course, strategy game training (active control) or movie screenings (passive control). We assessed cognitive performance in working memory, alertness, divided attention, and verbal fluency on a weekly basis. Trajectories were modeled using Generalized Additive Mixed Models to account for temporally limited transfer effects and intra-individual variation in cognitive performance. Our results provide no evidence of cognitive improvement differing between the Spanish and either of the control groups during any phase of the training period. We did, however, observe an effect of baseline cognition, such that individuals with low cognitive baselines increased their performance more in the L2 group than comparable individuals in the control groups. We discuss these findings against the backdrop of the cognitive training literature and Complex Dynamic Systems Theory.


Figure 1 Development of each group over time.
Figure 2 Interaction between year and program.
Communicative language teaching: Structure-Based or Dynamic Usage-Based?

July 2022

·

250 Reads

·

9 Citations

Journal of the European Second Language Association

Although communicative language teaching (CLT) was thought to have revolutionized classroom practice, there are “weak” and “strong” versions (Howatt, 1984). Most foreign language classrooms in the world still favor weak versions with structure-based (SB) views on language (Lightbown & Spada, 2013), and practice in the Netherlands is not much different (West & Verspoor, 2016). However, a small group of teachers in the Netherlands started teaching French as a second language with a strong CLT program in line with Dynamic Usage-Based (DUB) principles. Rather than focusing on rule learning and explicit grammar teaching to avoid errors, the DUB program takes the dynamics of second-language development into consideration and focuses on the three key elements of usage-based theory: frequency, salience and contingency. These translate into a great deal of exposure, repetition, learning the meaning of every single word through gestures, and presenting whole chunks of language, all without explicit grammar teaching. This study aims to compare the effects of the SB and DUB instructional programs after three years. We traced the second-language development of 229 junior high school students (aged 12 to 15) learning French in the Netherlands over three years. The participants took three oral tests over the course of three years (568 interviews) and wrote seven narratives on the same topic (1511 narratives). As expected, the DUB approach, which is in line with a strong CLT version, was more effective in achieving proficiency in both speaking and writing and equally effective in achieving accuracy.


Longitudinal analyses of depression and anxiety highlight greater prevalence during COVID-19 lockdowns in the Dutch general population and a continuing increase in suicidal ideation in young adults

May 2022

·

4 Reads

·

2 Citations

Objective The pandemic of the coronavirus disease 2019 (COVID-19) has led to an increased burden on mental health. This study therefore investigated the development of major depressive disorder (MDD), generalized anxiety disorder (GAD), and suicidal ideation in the Netherlands during the first fifteen months of the pandemic and three nation-wide lockdowns. Methods Participants of the Lifelines Cohort Study –a Dutch population-based sample-reported current symptoms of MDD and GAD, including suicidal ideation, according to DSM-IV criteria using a digital structured questionnaire. Between March 2020 and June 2021, 36,106 participants (aged 18-96) filled out a total of 629,811 questionnaires across 23 time points. Trajectories over time were estimated using generalized additive models and analyzed in relation to age, sex, and lifetime history of MDD/GAD to identify groups at risk. Results We found non-linear trajectories for MDD and GAD with a higher number of symptoms and prevalence rates during periods of lockdown. The point prevalence of MDD and GAD peaked during the third hard lockdown at 2.88% (95% CI: 2.71%–3.06%) and 2.92% (95% CI: 2.76%-3.08%), respectively, in March 2021. Women, younger adults, and participants with a history of MDD/GAD reported significantly more symptoms. For suicidal ideation, we found a linear increase over time in younger participants which continued even after the lockdowns ended. For example, 4.63% (95% CI: 3.09%-6.96%) of 20-year-old participants reported suicidal ideation at our last measured time point in June 2021, which represents a 4.14x increase since the start of the pandemic. Conclusions Our study showed greater prevalence of MDD and GAD during COVID-19 lockdowns suggesting that the pandemic and government enacted restrictions impacted mental health in the population. We furthermore found a continuing increase in suicidal ideation in young adults. This warrants for alertness in clinical practice and prioritization of mental health in public health policy.


Citations (17)


... Explainable predictions are essential for acceptance by (legal) experts. Schepers et al. (2024) predict citations in the context of Dutch case law and their method provides explainable predictions to legal experts in the form of meta-data. They report an explicit preference for models that predict false positives, i.e., that do not miss any positive sample. ...

Reference:

Predicting potentially unfair clauses in Chilean terms of services with natural language processing
Predicting citations in Dutch case law with natural language processing

Artificial Intelligence and Law

... We found associations between loneliness, depression and fatigue. Such relationships were also previously found in the general population due to the COVID-19 pandemic and the consequent lockdowns (Kalfas et al., 2024;Ori, Wieling, Lifelines Corona Research, & van Loo, 2023). Our results therefore not necessarily reflect an effect of having experienced SARS-CoV-2 infection but may also reflect impacts of the pandemic. ...

Longitudinal analyses of depression, anxiety, and suicidal ideation highlight greater prevalence in the northern Dutch population during the COVID-19 lockdowns

Journal of Affective Disorders

... The strong version of Communicative Language Teaching (CLT) advances the claim that language is acquired through communication rather than explicit instruction. Aligned with this perspective, Dynamic Usage-Based (DUB) principles emphasize meaningful exposure, repetition, and the integration of gestures and language chunks, enabling learners to develop proficiency without the need for explicit grammar instruction (Rousse-Malpat et al., 2022). This approach highlights the potential of authentic, communicative practices to stimulate the natural development of language systems. ...

Communicative language teaching: Structure-Based or Dynamic Usage-Based?

Journal of the European Second Language Association

... Cross-lingual transfer learning has emerged as a promising approach, where models pretrained on high-resource languages are fine-tuned for low-resource languages. This technique has shown effectiveness in various natural language processing tasks, including task-oriented dialogue systems, document representation, and part-of-speech tagging (Fuad & Al-Yahya, 2022;Gong et al., 2021;Vries et al., 2022). ...

Make the Best of Cross-lingual Transfer: Evidence from POS Tagging with over 100 Languages
  • Citing Conference Paper
  • January 2022

... Atualmente, o processamento da linguagem natural (PLN) é o método de inteligência artificial mais habitualmente usado na prática do direito (Haney, 2020). Sobretudo, ao considerar que o Direito é fortemente baseado na palavra escrita, é um dos campos que podem se beneficiar dessa abordagem (Aletras et al., 2016;Ikram & Chakir, 2019;Katz, Bommarito & Blackman, 2017;Kowsrihawat, Vateekul, & Boonkwan, 2018;Long et al., 2019;Şulea et al., 2017a;Şulea et al., 2017b;Virtucio et al., 2018;Medvedeva, Wieling, & Vols, 2023). Trata-se de uma técnica chave para mineração de textos, normalmente, participando do pré-processamento dos dados, transformando os textos em números(Aranha & Passos, 2006). ...

Rethinking the field of automatic prediction of court decisions

Artificial Intelligence and Law

... 1 Older adults in general Decline in cognitive functions (Brasser et al., 2022;Dziemian et al., 2021;Kliesch et al., 2022;Krebs et al., 2021;Najberg et al., 2021;Schättin et al., 2019;Seinsche et al., 2023;Studer-Luethi et al., 2021Tinello et al., 2023;Zuber et al., 2021) Decline in physical abilities and physical health (Adcock et al., 2019;Adcock et al., 2020a, Adcock et al., 2020bNeumann et al., 2018;Ringgenberg et al., 2022;Schättin et al., 2019 ;Seinsche et al., 2023) Self-stigma (Adcock et al., 2019) and public stigma (Adcock et al., 2019;Neumann et al., 2018) Impaired mental health (Krebs et al., 2021;Najberg et al., 2021;Seinsche et al., 2023) 2 Pre-frail older adults Frailty risk state associated with one or two of the following criteria: unintentional weight loss; weakness or poor handgrip strength; selfreported exhaustion; slow walking speed; and low physical activity (Belleville et al., 2020(Belleville et al., , 2023 Public stigma (Belleville et al., 2023) 3 Mobility impaired older adults ...

Cognitive Benefits of Learning Additional Languages in Old Adulthood? Insights from an Intensive Longitudinal Intervention Study
  • Citing Article
  • August 2022

Applied Linguistics

... Additionally, several works have developed NLP models to classify legal documents. For instance, they have created automated systems to identify, categorize, and forecast court decisions [12,13,14,15,16], or have been used to forecast future citations to case law [17,18] and to summarize legal cases using large language models (LLM) [19,20,21,22,23]. ...

Automatically Identifying Eviction Cases and Outcomes Within Case Law of Dutch Courts of First Instance

... To address this issue, researchers have explored methods to preserve as much of the original model's embeddings as possible, especially when the source and target tokenizers share morphosemantic similarities (Artetxe et al., 2020;Garcia et al., 2021;Gogoulou et al., 2022). However, the application of these methods is limited by the availability of shared tokens and the morphosemantic proximity between the involved languages (de Vries et al., 2021). ...

Adapting Monolingual Models: Data can be Scarce when Language Similarity is High
  • Citing Conference Paper
  • January 2021

... Additionally, several works have developed NLP models to classify legal documents. For instance, they have created automated systems to identify, categorize, and forecast court decisions [12,13,14,15,16], or have been used to forecast future citations to case law [17,18] and to summarize legal cases using large language models (LLM) [19,20,21,22,23]. ...

Automatic Judgement Forecasting for Pending Applications of the European Court of Human Rights